GIF Thumbnails: Attract More Clicks to Your Videos

نویسندگان

چکیده

With the rapid increase of mobile devices and online media, more people prefer posting/viewing videos online. Generally, these are presented on video streaming sites with image thumbnails text titles. While facing huge amounts videos, a viewer clicks through certain high probability because its eye-catching thumbnail. However, current created manually, which is time-consuming quality-unguaranteed. And static contain very limited information corresponding prevents users from successfully clicking what they really want to view. In this paper, we address novel problem, namely GIF thumbnail generation, aims automatically generate for consequently boost their Click-Through-Rate (CTR). Here, an animated file consisting multiple segments video, containing target than To support study, build first benchmark dataset that consists 1070 covering total duration 69.1 hours, 5394 manually-annotated GIFs. solve propose learning-based automatic generation model, called Generative Variational Dual-Encoder (GEVADEN). As not relying any user interaction (e.g. time-sync comments real-time view counts), model applicable newly-uploaded/rarely-viewed videos. Experiments our built show GEVADEN significantly outperforms several baselines, including video-summarization highlight-detection based ones. Furthermore, develop pilot application proposed platform 9814 1231 shows achieves 37.5% CTR improvement over traditional thumbnails. This further validates effectiveness promising prospect

برای دانلود باید عضویت طلایی داشته باشید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Get more Clicks!

Sponsored search has become increasingly important due to the rapid development of Web search engines and pay per click (PPC) is amongst one of the most important advertising models search engines currently use. One of the key questions in sponsored search is that: Given a query or a substituted keyword, which ads should search engines display to the users in order to maximize their revenue? In...

متن کامل

For Your Eyes Only: Learning to Summarize First-Person Videos

With the increasing amount of video data, it is desirable to highlight or summarize the videos of interest for viewing, search, or storage purposes. However, existing summarization approaches are typically trained from third-person videos, which cannot generalize to highlight the first-person ones. By advancing deep learning techniques, we propose a unique network architecture for transferring ...

متن کامل

More than Paradoxes to Offer: Exploring Motivations to Attract Women to Computing

The question we raise in this paper is why some women choose not to go into computing. What motivates their reasons for selecting their major and occupational goals? We interviewed twelve first year women who are not majoring in Computer Science and found half had pro-social goals for their careers. When asked if participants thought that people working in computing are helping people, some tol...

متن کامل

Does warmer China land attract more super typhoons?

Accurate prediction of where and when typhoons (or named hurricanes which form over the North Atlantic Ocean) will make landfall is critical to protecting human lives and properties. Although the traditional method of typhoon track prediction based on the steering flow theory has been proven to be an effective way in most situations, it slipped up in some cases. Our analysis of the long-term Ch...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

ژورنال

عنوان ژورنال: Proceedings of the ... AAAI Conference on Artificial Intelligence

سال: 2021

ISSN: ['2159-5399', '2374-3468']

DOI: https://doi.org/10.1609/aaai.v35i4.16416